A story of false data
In fiction, cancer research is about trying to find a cure for cancer. A single cure, that fixes everybody's cancer. While this certainly makes for good stories, and would indeed be wonderful if anyone ever managed to come up with it, most real-life cancer researchers settle for a lesser goal that is still an achievement: find drugs that will slow down (or, in some cases, remove) a particular type of cancer in a particular type of person. For example, they might come up with research that only works on nose cancer in Asians, and then only works some of the time. Nevertheless this can still add years of life to some people and is therefore worth checking.
In 2013-2014 I was involved in therapeutic aspects of cancer research, in further investigation of the interplay between the PRDM1 gene and the NF-κB protein that leads to different susceptibility rates of NK/T cell lymphoma (NKTCL) cancer patients towards chemotherapeutic agents. The research team of my supervisor had a leading study using a 26S proteasome inhibitor drug, bortezomib (Velcade), to act against NK/T cell lymphoma.1 With the assumption that IL2 is an important factor in sustaining NK/T cell lymphoma, we investigated the combination treatment that can reduce bortezomib dosage by using either of the two fusion anti-IL2 receptor antibodies, basiliximab (Simulect®) and denileukin diftitox (ONTAK®). We found that IC50 (median inhibitory concentration) of NKTCL cells being treated by bortezomib can be decreased by ~53% to ~90% when combined with a low level of basiliximab. This was a promising result, but it still needed testing "in vivo" (in a live setting). As the lab I was in did not have enough live sample (and my supervisor was retiring and the grants were running out), we came up the idea of asking another lab (one in mainland China with whom we had long been in collaboration) to run the live tests for us.
I visited China, trained the people there to perform a couple of in vivo tests, and finished the job. Several months later I married and moved to the UK, and the lab in China kept informing us about the progress of their study by sending us results in pictures. Those results were coherent with the in vitro test results we had previously found in our lab. Maybe we really did have a new, more effective chemotherapy drug on our hands for people with that particular cancer type? I had to fly back to Hong Kong and check things out (we tried working remotely but the lab really wanted me there). Things seemed to be playing out like a film script, aeroplanes and all.
But sadly, our further analysis showed that the data from the China lab was very likely to be fake. I found two pictures sent at two different times, supposedly representing different samples, were actually identical. By reflecting and rotating one of the pictures later sent to me, I could not see any difference with the one that had been sent earlier. We decided not to publish it after all.
So, does our combination work? It has not been disproven. But science does not work on not disproving things, it works on positive proof. It might still be possible for other labs to re-check our combination, and we hope they will do so from our partial publications, but we were not able to show it ourselves. And we ended up being a little annoyed with that lab for faking their data (perhaps because they couldn't be bothered to run the real tests?) and causing all that excitement for nothing. But these things happen.
The quest to improve the lives of people with cancer continues, one small uncertain arduous step at a time.
References
1. L. Shen, W.Y. Au, T. Guo, et al. Proteasome inhibitor bortezomib-induced apoptosis in natural killer (NK)-cell leukemia and lymphoma: an in vitro and in vivo preclinical evaluation. Blood, 2007 Jul 1;110(1):469-70.
In fiction, cancer research is about trying to find a cure for cancer. A single cure, that fixes everybody's cancer. While this certainly makes for good stories, and would indeed be wonderful if anyone ever managed to come up with it, most real-life cancer researchers settle for a lesser goal that is still an achievement: find drugs that will slow down (or, in some cases, remove) a particular type of cancer in a particular type of person. For example, they might come up with research that only works on nose cancer in Asians, and then only works some of the time. Nevertheless this can still add years of life to some people and is therefore worth checking.
In 2013-2014 I was involved in therapeutic aspects of cancer research, in further investigation of the interplay between the PRDM1 gene and the NF-κB protein that leads to different susceptibility rates of NK/T cell lymphoma (NKTCL) cancer patients towards chemotherapeutic agents. The research team of my supervisor had a leading study using a 26S proteasome inhibitor drug, bortezomib (Velcade), to act against NK/T cell lymphoma.1 With the assumption that IL2 is an important factor in sustaining NK/T cell lymphoma, we investigated the combination treatment that can reduce bortezomib dosage by using either of the two fusion anti-IL2 receptor antibodies, basiliximab (Simulect®) and denileukin diftitox (ONTAK®). We found that IC50 (median inhibitory concentration) of NKTCL cells being treated by bortezomib can be decreased by ~53% to ~90% when combined with a low level of basiliximab. This was a promising result, but it still needed testing "in vivo" (in a live setting). As the lab I was in did not have enough live sample (and my supervisor was retiring and the grants were running out), we came up the idea of asking another lab (one in mainland China with whom we had long been in collaboration) to run the live tests for us.
I visited China, trained the people there to perform a couple of in vivo tests, and finished the job. Several months later I married and moved to the UK, and the lab in China kept informing us about the progress of their study by sending us results in pictures. Those results were coherent with the in vitro test results we had previously found in our lab. Maybe we really did have a new, more effective chemotherapy drug on our hands for people with that particular cancer type? I had to fly back to Hong Kong and check things out (we tried working remotely but the lab really wanted me there). Things seemed to be playing out like a film script, aeroplanes and all.
But sadly, our further analysis showed that the data from the China lab was very likely to be fake. I found two pictures sent at two different times, supposedly representing different samples, were actually identical. By reflecting and rotating one of the pictures later sent to me, I could not see any difference with the one that had been sent earlier. We decided not to publish it after all.
So, does our combination work? It has not been disproven. But science does not work on not disproving things, it works on positive proof. It might still be possible for other labs to re-check our combination, and we hope they will do so from our partial publications, but we were not able to show it ourselves. And we ended up being a little annoyed with that lab for faking their data (perhaps because they couldn't be bothered to run the real tests?) and causing all that excitement for nothing. But these things happen.
The quest to improve the lives of people with cancer continues, one small uncertain arduous step at a time.
References
1. L. Shen, W.Y. Au, T. Guo, et al. Proteasome inhibitor bortezomib-induced apoptosis in natural killer (NK)-cell leukemia and lymphoma: an in vitro and in vivo preclinical evaluation. Blood, 2007 Jul 1;110(1):469-70.
No comments:
Post a Comment